Awk Program to Determine Maximum

Question

I'm trying to determine the maximum values of columns S2 & S4 but only for the corresponding S1 columns that have BBB. The output would be S2 max=800 and S4 max=90. Currently have: awk 'BEGIN { FS = " " ; OFS = " "; } { if(NR%3==0){max=$3} { s2+=substr($2,1,3); s4+=substr($4,1,2) } } ; { $3 >= max } END { print "Max S2 = " s2, "; Max S4 = " s4 } ' file.txt

   S1 S2 S3 S4 
   --- --- - --
   AAA 100 A 99
   BBB 200 B 90
   CCC 300 C 80
   AAA 400 A 70
   BBB 500 B 60
   CCC 600 C 50
   AAA 700 A 40
   BBB 800 B 30
   CCC 900 C 20
   AAA 999 A 10

Answer 1

The shell you are using should not make any difference, it doesn't change the way awk behaves. It may change the way you provide variable input to awk, but your problem here doesn't require any.

Here you can simply identify the records (lines) containing BBB and then keep track of the maximum values for field-2 and field-4 and then output the results using the END rule. You can use a traditional if or a ternary operator as @RavinderSingh13 has used. A traditional if would be:

awk '/BBB/{
    if ($2 > s2)
        s2 = $2
    if ($4 > s4)
        s4 = $4
}
END{
    printf "S2 max=%d and S4 max=%d\n", s2, s4
}' file

Example Use/Output

With your data in the filename file, you can just change to the directory containing file (make sure you change file to whatever your input file name is) and then select-copy above and middle-mouse-paste into the terminal to test, e.g.

$ awk '/BBB/{
>     if ($2 > s2)
>         s2 = $2
>     if ($4 > s4)
>         s4 = $4
> }
> END{
>     printf "S2 max=%d and S4 max=%d\n", s2, s4
> }' file
S2 max=800 and S4 max=90